Project Summary The purpose of this proposal is to provide training opportunities and mentorship in latent variable and dyadic data analyses. These skills will be used to examine the science that underlies parent-child mealtime interactions and to predict weight gain in toddlers. This training grant includes coursework in advanced statistical modeling and mentorship geared towards professional development. This coursework and mentorship provides the opportunity to build a foundation as an independent research scientist. The pediatric obesity epidemic contributes to health disparities by increasing the risk for health outcomes including cardiovascular disease, asthma, and obstructive sleep disorders. Obesity occurs disproportionately among low-income and minority children. The origins of pediatric obesity begin early in life; eating habits formed during early childhood comprise the basis of long term lifestyle behaviors. Parent-child mealtime interactions are linked to eating and overweight in children. Furthermore, disordered maternal eating attitudes often translate to less harmonious mealtime interactions. Therefore, investigation into bidirectional parent-child mealtime interactions may yield insight into disordered eating patterns and ultimately reduce obesity risk. The proposed study seeks to examine the relation between maternal eating attitudes and childhood health outcomes. Specifically, the study aims to create a theoretically inclusive model to determine if mealtime interactions mediate the relation between maternal eating attitudes and child weight over time. Latent variable modeling techniques will be used to examine if maternal eating attitudes predict mealtime behaviors and toddler weight trajectory. Additionally, the proposed study aims to elucidate the bidirectional relations between parent-child interactions during mealtimes. This will be accomplished using analysis of intensive micro-level dyadic longitudinal data. The proposed study is unique in the use of observational methods to examine longitudinal dyadic interactions. This project is in line with NHLBI's mission to prevent HLBS diseases through identification of changeable risk factors associated with obesity in young children.